tf.fake_quant_with_min_max_args_gradient

tf.fake_quant_with_min_max_args_gradient

tf.fake_quant_with_min_max_args_gradient

fake_quant_with_min_max_args_gradient(
    gradients,
    inputs,
    min=None,
    max=None,
    num_bits=None,
    name=None
)

Defined in tensorflow/python/ops/gen_array_ops.py.

See the guide: Tensor Transformations > Fake quantization

Compute gradients for a FakeQuantWithMinMaxArgs operation.

Args:

• gradients: A Tensor of type float32. Backpropagated gradients above the FakeQuantWithMinMaxArgs operation.
• inputs: A Tensor of type float32. Values passed as inputs to the FakeQuantWithMinMaxArgs operation.
• min: An optional float. Defaults to -6.
• max: An optional float. Defaults to 6.
• num_bits: An optional int. Defaults to 8.
• name: A name for the operation (optional).

Returns:

A Tensor of type float32. Backpropagated gradients below the FakeQuantWithMinMaxArgs operation: gradients * (inputs >= min && inputs <= max).